honcho

Automate cross-session memory management and user modeling for Hermes via Honcho.

Updated Jun 28, 2026
One-click install
npx skills add https://github.com/jleechanorg/hermes-agent --skill honcho-jleechanorg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: honcho
Source: https://github.com/jleechanorg/hermes-agent/tree/main/optional-skills/autonomous-ai-agents/honcho
Command: npx skills add https://github.com/jleechanorg/hermes-agent --skill honcho-jleechanorg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Honcho provides AI-native cross-session memory modeling and persistent context for Hermes, allowing a shared user representation across profiles while giving each Hermes profile its own AI peer for isolated reasoning and personalized interactions.

Core Features & Use Cases

  • Cross-session memory and user modeling: shared workspace with per-profile AI peers and individualized observations.
  • Context injection and recall controls: configurable recall modes, observation toggles, and memory budgets to balance detail and tokens.
  • Multi-profile orchestration: seamless handling of multiple Hermes profiles sharing one workspace, with isolated AI peers and consolidated user context.
  • Use Case: maintain a consistent user representation across sessions while enabling each profile to reason with its own AI identity.

Quick Start

Install Honcho via Hermes and run hermes honcho setup to enable cross-session memory and per-profile AI peers.

Frequently Asked Questions about honcho

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I enable cross-session memory for Hermes profiles?

Cross-session memory for Hermes profiles is enabled by installing Honcho via Hermes and running the setup command. This activates the persistent memory API, per-profile AI peers, and shared user context across sessions.

What is cross-session user modeling and how does it work with multiple AI peers?

Cross-session user modeling maintains a shared workspace of persistent context for a single user. It assigns each Hermes profile an isolated AI peer for independent reasoning while consolidating observations into one unified user representation.

Can I control memory recall and observation toggles to manage token budgets?

Yes, Honcho provides configurable recall modes, observation toggles, and memory budgets. These controls let you balance the level of contextual detail injected against token usage for each session.

Does Honcho support self-hosted deployments for Hermes?

Honcho supports both cloud and self-hosted deployments for Hermes. It handles per-profile configuration and session-scoped recall through the memory API across these deployment types.

Why use per-profile AI peers instead of a single shared context?

Per-profile AI peers allow each Hermes profile to maintain its own isolated reasoning identity and personalized interactions. This prevents context bleed between profiles while still consolidating shared user context in the background.